Özet:

In recent years, various tools and algorithms have been proposed and continue to be proposed by researchers to develop highly successful medical decision support systems. However, the clinical use of these algorithms is very limited due to various limitations. Making the necessary software installations to run the algorithm, lack of programming knowledge are some of these restrictions. In this study, a web-based classification software developed with the Julia programming language, which can be used by physicians in their medical research and clinical decisions, is introduced. Through this software, coronary artery disease detection was performed with the Cleveland heart disease database, which is a publicly accessible data set. The dataset was classified with eight different classifiers (KNN, SVM, DT, RF, AdaBoost, Gauss Naive Bayes, LDA, LR) supported by the software. The metrics obtained by 10-fold cross-validation of the data set are reported. The SVM classifier achieved the highest classification accuracy with 86.44%. The software proposed in this study may assist clinicians in research and patient identification.

Anahtar Kelimeler: